The primary aim of this study is to determine the association of cognitive function and the incidence and prevalence of Alzheimer's disease (AD) with use of estrogen replacement therapy (ERT) or combined estrogen/progestin therapy (HRT) in a group of women whose ERT/HRT use can be characterized based on computer-stored data and medical records. The investigators will also examine the association of cognitive function and AD with apoE genotype, and the interactions between ERT/HRT, apoE genotype and cognitive function and AD. Secondarily, the study will examine the association of other reproductive variables with cognitive function and apoE genotype. The study will use computer-stored prescription data and chart review to assemble a study population comprised of women who are 75+ years of age and are known to be current users of ERT/HRT (n=2,000), past users of ERT/HRT (n=1,000), or never users of ERT/HRT (n=2,000), screen them for possible dementia using the modified Telephone Interview for Cognitive Status, and assess apoE genotype. Women with possible dementia will be categorized as having AD, non-AD dementia, or as non-demented based on additional tests of cognitive function and review of medical records. An analysis of the prevalence odds ratio for AD in relation to ERT/HRT will be done. A random sample of current and never users of ERT/HRT who are non-demented based on the screening examination (n=120 in each group) will undergo a battery of standardized neuropsychologic tests that assess cognitive domains reported to be affected by estrogen in prior studies, as well as domains not previously associated with ERT/HRT use. Detailed information on reproductive history will be collected from these women in structured interviews. An analysis of cognitive function comparing current and never users of ERT/HRT and reproductive factors in these women will be done. Current and never users of ERT/HRT who are not demented at the time of the baseline study will be followed for an average of 3 years to ascertain incident cases of AD. An analysis comparing the incidence of AD in current and never users of ERT/HRT at baseline will be performed.